Biostatistical Models for Predicting Mortality in Intensive Care Units

Background: Technology has transformed mortality prediction in the ICU through its ability to make more accurate and timely decisions based on data. With the ongoing development of these technologies, they become very promising in increasing the survival rates and promoting the individual care of critical care. Methods: In current study, we have aimed to examine the role of biostatistical models in predicting mortality in ICU within hospitals and health centers in the West Bank in terms of (Input Variables, Model Type / Methodology, Calibration and Discrimination, Temporal Resolution, Handling of Missing Data, Interpretability, External Validation and Clinical Integration). (223) physicians responded to an online questionnaire. SPSS was used to screen and analyze collected primary data. Results: Results indicated the acceptance of our hypothesis “biostatistical models can be used for predicting mortality in Intensive Care Units from perspective of physicians”. Among the chosen variables, it was noticed that model type/methodology with b= 0.256 scored the highest in influence on mortality rate prediction. Conclusion: Study recommended that hospitals should integrate explainable models like models that use SHAP values into electronic health record systems to enable real-time mortality predictions in the intensive care units. Trial Registration: J-2025/A 95/N.